Overclock.net › Forums › Intel › Intel CPUs › [Official] Delidded Club / Guide
New Posts  All Forums:Forum Nav:

[Official] Delidded Club / Guide - Page 2462

post #24611 of 33658
Quote:
Originally Posted by Wirerat View Post

Quote:
Originally Posted by deepor View Post

People report both, CLP better or CLU better. What's better might change depending on the temperatures. It might depend on how it was applied exactly, the amount of it, how the cooler is mounted etc., random differences.

I think somewhere I've seen the numbers Coolaboratory themselves mention for thermal conductivity of their products and CLU had slightly better numbers.
the delta is so small between the two when compared to the 25c reduction vs stock.

Thank you both. I am going to order some ultra. I couldnt find my tube when I was cleaning my loop, and ended up putting EK tim that comes with the GPU blocks on it. Temps are about 10C higher, and I had to bump voltage +.012 to stabilize again. Voltage/Temps are a dirty little cycle.
Blue540
(14 items)
 
x99
(18 items)
 
B.M.O.
(10 items)
 
CPUMotherboardGraphicsGraphics
6800k ASUS X99-A EVGA 980 FTW ACX 2.0 EVGA 980 FTW ACX2.0 
RAMHard DriveHard DriveCooling
Corsair Vengeance CMK32GX4M4B3200C16W Crucial MX100 WD Blue 1TB EK Supremacy EVO 
CoolingCoolingCoolingOS
EK 480 PE EK 240 PE EK D5 Vario Windows 10 
MonitorKeyboardPowerCase
Samsung 49KU7500 (curved 4k) Corsair K65 NZXT Hale 90 V2 1000W Thermaltake X9 
MouseMouse Pad
Logitch G100S X-Trac Ripper 
CPUMotherboardGraphicsRAM
6400k msi A75MA-P33 Sapphire 6670 passive hynix 2 GB dual channel 
Hard DriveOSMonitorKeyboard
WD  windows 8.1 e-machine wireless 
PowerCase
king win 600w custom built 
  hide details  
Reply
Blue540
(14 items)
 
x99
(18 items)
 
B.M.O.
(10 items)
 
CPUMotherboardGraphicsGraphics
6800k ASUS X99-A EVGA 980 FTW ACX 2.0 EVGA 980 FTW ACX2.0 
RAMHard DriveHard DriveCooling
Corsair Vengeance CMK32GX4M4B3200C16W Crucial MX100 WD Blue 1TB EK Supremacy EVO 
CoolingCoolingCoolingOS
EK 480 PE EK 240 PE EK D5 Vario Windows 10 
MonitorKeyboardPowerCase
Samsung 49KU7500 (curved 4k) Corsair K65 NZXT Hale 90 V2 1000W Thermaltake X9 
MouseMouse Pad
Logitch G100S X-Trac Ripper 
CPUMotherboardGraphicsRAM
6400k msi A75MA-P33 Sapphire 6670 passive hynix 2 GB dual channel 
Hard DriveOSMonitorKeyboard
WD  windows 8.1 e-machine wireless 
PowerCase
king win 600w custom built 
  hide details  
Reply
post #24612 of 33658
Quote:
Originally Posted by Valgaur View Post

sorry sometimes feels like peeps don't read the stuffs anymore tongue.gif

Np smile.gif and saddly that's true. Though I admit I may have lost some patience myself, it is now more than one year that I'm planning my build haha ( and doing it on a really crappy comp mad.gif ) I didn't even know what a cpu is x) .. Really want to be part of it ! post my numbers here and on ocn, and do all those things biggrin.gif You guys are inspiring ^_^

Okay so I will do direct-to-die cooling with CLU and an EK copper block and its PreciseMount kit.. since I'm headed towards high-end cooling ( Phobya 1260 for cpu + SLI ) might as well go full for it, as direct-die will yield better temps. Even if it could be hard to clean the CLU off of the copper block afterwards, that's no biggie as I want to stick with copper blocks within my loop.
I'll just be careful as hell when processing.
post #24613 of 33658
Quote:
Originally Posted by White Wind View Post

Okay so I will do direct-to-die cooling with CLU and an EK copper block and its PreciseMount kit..

I wish someone would have made this suggestion before I went direct die-->CLU :

Go with a nickel block. The pricing is only a little more, you will not experience the same staining that many of us here have. The CLU is almost impossible to get off of copper, I had to sand mine off to get it clean. I dont believe you will notice any temperature difference, but it might be worth not having the cleaning headache later.
Blue540
(14 items)
 
x99
(18 items)
 
B.M.O.
(10 items)
 
CPUMotherboardGraphicsGraphics
6800k ASUS X99-A EVGA 980 FTW ACX 2.0 EVGA 980 FTW ACX2.0 
RAMHard DriveHard DriveCooling
Corsair Vengeance CMK32GX4M4B3200C16W Crucial MX100 WD Blue 1TB EK Supremacy EVO 
CoolingCoolingCoolingOS
EK 480 PE EK 240 PE EK D5 Vario Windows 10 
MonitorKeyboardPowerCase
Samsung 49KU7500 (curved 4k) Corsair K65 NZXT Hale 90 V2 1000W Thermaltake X9 
MouseMouse Pad
Logitch G100S X-Trac Ripper 
CPUMotherboardGraphicsRAM
6400k msi A75MA-P33 Sapphire 6670 passive hynix 2 GB dual channel 
Hard DriveOSMonitorKeyboard
WD  windows 8.1 e-machine wireless 
PowerCase
king win 600w custom built 
  hide details  
Reply
Blue540
(14 items)
 
x99
(18 items)
 
B.M.O.
(10 items)
 
CPUMotherboardGraphicsGraphics
6800k ASUS X99-A EVGA 980 FTW ACX 2.0 EVGA 980 FTW ACX2.0 
RAMHard DriveHard DriveCooling
Corsair Vengeance CMK32GX4M4B3200C16W Crucial MX100 WD Blue 1TB EK Supremacy EVO 
CoolingCoolingCoolingOS
EK 480 PE EK 240 PE EK D5 Vario Windows 10 
MonitorKeyboardPowerCase
Samsung 49KU7500 (curved 4k) Corsair K65 NZXT Hale 90 V2 1000W Thermaltake X9 
MouseMouse Pad
Logitch G100S X-Trac Ripper 
CPUMotherboardGraphicsRAM
6400k msi A75MA-P33 Sapphire 6670 passive hynix 2 GB dual channel 
Hard DriveOSMonitorKeyboard
WD  windows 8.1 e-machine wireless 
PowerCase
king win 600w custom built 
  hide details  
Reply
post #24614 of 33658
Quote:
Originally Posted by inedenimadam View Post

I wish someone would have made this suggestion before I went direct die-->CLU

Well, I think that was curiosity and..just luck that made me know about that... rather lucky on this one tongue.gif
Quote:
Originally Posted by inedenimadam View Post

Go with a nickel block.
you will not experience the same staining that many of us here have
it might be worth not having the cleaning headache later.

Well, if it's just about getting stains and having to sand, I don't think that would bother me that much, plus I won't be dismounting the block from the cpu until I've decided to upgrade it ( or should I, for maintenance purpose ? )

You said you can't tell if yours was CLP or CLU.. maybe that was CLP which is known to be the one hard to clean ?

And you're probably right about temps difference between copper and nickel, I can't tell myself, that's just that it is said that copper is a better conductor, and if I chose nickel I would have to go nickel on my gpu's too.. maybe the few degrees will be there when considering the whole loop.. I don't know

EDIT: oh, the "suggestion" you were talking about is the one you made about nickel block lol.. okay got it now x)
Edited by White Wind - 3/26/14 at 5:36pm
post #24615 of 33658
Quote:
Originally Posted by White Wind View Post

Well, I think that was curiosity and..just luck that made me know about that... rather lucky on this one tongue.gif
Well, if it's just about getting stains and having to sand, I don't think that would bother me that much, plus I won't be dismounting the block from the cpu until I've decided to upgrade it ( or should I, for maintenance purpose ? )

You said you can't tell if yours was CLP or CLU.. maybe that was CLP which is known to be the one hard to clean ?

And you're probably right about temps difference between copper and nickel, I can't tell myself, that's just that it is said that copper is a better conductor, and if I chose nickel I would have to go nickel on my gpu's too.. maybe the few degrees will be there when considering the whole loop.. I don't know

EDIT: oh, the "suggestion" you were talking about is the one you made about nickel block lol.. okay got it now x)

Just gonna throw my 2cents.gif in here:

The reason that BOTH CLU and CLP stain the copper is that the Gallium (that's the metal) in the CLP/U is actually absorbed by the copper. In time (and it's not a consistent amount of time, as the different alloys of copper used will absorb the gallium at different rates), the gallium is no longer actually making a proper bond as a TIM between the two surfaces and the temps will start to rise because of it. So from my own experiences with this, I will only use a "normal" TIM when doing a direct die mount and an unplated copper block, and in all honesty, i do the same for the plated blocks if direct die. I would only recommend the Coolabs TIMs if you are going to put the IHS back on, not for a naked die. Going naked die will benefit you more temps than using the CLP/U as a TIM in this situation, in naked die the CLP/U might net you an extra 1c, maybe 2c. But for how long?

And remember, the block is copper, even if it's nickel plated, it's still copper, the plating is VERY thin, so there isn't a temp difference between the plain copper block and the nickel plated copper block.

thumb.gif
Edited by Jimhans1 - 3/26/14 at 11:01pm
post #24616 of 33658
Good post thanks ! I wasn't aware at all that the TIM bond could suffer as well. So okay, no full copper + CLU on a naked die.
Quote:
Originally Posted by Jimhans1 View Post

Going naked die will benefit you more temps than using the CLP/U as a TIM in this situation

There, you mean that going "naked-die + grease" will get me higher temps than "naked-die + CLU", right ?..( I'm not a native english speaker sorry )
Or maybe you meant "naked-die + grease" not as good as "IHS on + CLU".. I wonder about it now.
Quote:
Originally Posted by Jimhans1 View Post

And remember, the block is copper, even if it's nickel plated, it's still copper, the plating is VERY thin, so there isn't a temp difference between the plain copper block and the nickel plated copper block.

thumb.gif

Ow I thought that when we were talking about nickel blocks it was about full nickel blocks... so plated blocks are not plated just on the upside lol.. okay got it x)
That may be the way to go for me then. But in my radiator, the tubes will be made from copper, do I risk any corrosion if I opt for nickel-plated copper blocks ?
Now to see if EK make nickel-plated copper blocks, I'm to check that
post #24617 of 33658
Quote:
Originally Posted by White Wind View Post

Good post thanks ! I wasn't aware at all that the TIM bond could suffer as well. So okay, no full copper + CLU on a naked die.
There, you mean that going "naked-die + grease" will get me higher temps than "naked-die + CLU", right ?..( I'm not a native english speaker sorry )
Or maybe you meant "naked-die + grease" not as good as "IHS on + CLU".. I wonder about it now. Warning: Spoiler! (Click to show)
Ow I thought that when we were talking about nickel blocks it was about full nickel blocks... so plated blocks are not plated just on the upside lol.. okay got it x)
That may be the way to go for me then. But in my radiator, the tubes will be made from copper, do I risk any corrosion if I opt for nickel-plated copper blocks ?
Now to see if EK make nickel-plated copper blocks, I'm to check that

Naked die + normal TIM/grease will give temps maybe 1-2C higher than naked+CLP/U. Naked die itself with any TIM is only a few degrees C cooler than IHS+ CLU/P.

However, iff using the ihs, clu/p will give better temps (~5-20C) than using normal TIM on die without suffering from "pump out effect". thumb.gif
Edited by MrStrat007 - 3/27/14 at 5:04am
First Rig
(15 items)
 
  
Reply
First Rig
(15 items)
 
  
Reply
post #24618 of 33658
Quote:
Originally Posted by MrStrat007 View Post

Naked die + normal TIM/grease will give temps maybe 1-2C higher than naked+CLP/U. Naked die itself with any TIM is only a few degrees C cooler than IHS+ CLU/P.

However, iff using the ihs, clu/p will give better temps (~5-20C) than using normal TIM on die without suffering from "pump out effect". thumb.gif

I couldn't hope for a better input lol.. thank you ^_^ I have a couple more questions if you don't mind

So normal TIM on naked die can too be pumped out when in contact with copper ?

And if I go naked die + CLU + nickel-plated block ( so no pump out effect there ), would my loop suffer from corrosion? as the tubes in my radiator will be made from copper..
If yes and if I go that way, would there be something I could add in the loop to prevent/counteract that corrosion ?

Which way would you go ?lol ..I'll just take on your advice and be done with it ^_^ ..and stop the endless talk haha. But nevertheless big thanks ! ( again ) it is appreciated
post #24619 of 33658
That "pump out" stuff has nothing to do with copper or nickel or whatever. It's just that some thermal paste products don't stay where you've put it on such a tiny area like the die. The temperature changes and pressure make it move away. It gets squeezed out at the sides over time. You need to use a paste where people reported they had no problems. I think Arctic MX-4 might be good? or Noctua NT-H1? I don't remember really.
post #24620 of 33658
Quote:
Originally Posted by White Wind View Post

I couldn't hope for a better input lol.. thank you ^_^ I have a couple more questions if you don't mind

So normal TIM on naked die can too be pumped out when in contact with copper ?

And if I go naked die + CLU + nickel-plated block ( so no pump out effect there ), would my loop suffer from corrosion? as the tubes in my radiator will be made from copper..
If yes and if I go that way, would there be something I could add in the loop to prevent/counteract that corrosion ?

Which way would you go ?lol ..I'll just take on your advice and be done with it ^_^ ..and stop the endless talk haha. But nevertheless big thanks ! ( again ) it is appreciated

Quote:
Originally Posted by deepor View Post

That "pump out" stuff has nothing to do with copper or nickel or whatever. It's just that some thermal paste products don't stay where you've put it on such a tiny area like the die. The temperature changes and pressure make it move away. It gets squeezed out at the sides over time. You need to use a paste where people reported they had no problems. I think Arctic MX-4 might be good? or Noctua NT-H1? I don't remember really.

Like Deepor said, Pump-out is usually seen when using normal TIM or grease between the die and IHS. It is caused by the clamping pressure and thermal cycling of the cpu causing minute expansion/contractions of the TIM itself. Over time, this causes the TIM to shift, effectively "pumping out" from on top of the die, reducing contact with the IHS. This isn't caused by any particular metal, and if you wait for a few of the guys who run naked die, I'm sure they will post with their favourite TIM's for naked die operation and their experiences with them.

iirc The only corrosion you would get with naked+CLU+nickel would be if the CLU got through the plating to the copper base, at which point you have the same thing happen as using a pure copper block. There are corrosion inhibitors for your loop, such as Mayhems product and various standalone options, BUT they are more to prevent the effects of mixed metals in the same loop. Even if you used an all-copper cpu block, CLU would not affect your loop, only your block.

However, given the tiny temperature differences, if you choose to go bare die I would just use a regular TIM like MX4 of NT-H1, etc. and not worry about using the CLU/P. More peace of mind, basically the same performance, and less to worry about thumb.gif
First Rig
(15 items)
 
  
Reply
First Rig
(15 items)
 
  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel CPUs
Overclock.net › Forums › Intel › Intel CPUs › [Official] Delidded Club / Guide